Wolff-Parkinson-White syndrome is a cardiac condition characterized by abnormality in the heart’s system. The detection of this syndrome involves pattern detection from electrocardiogram images. It can be extremely life-threatening, due to which its detection at the right time is a crucial task. Various studies show that integrating recent revolutionary concepts like deep learning with the field of bioinformatics can be a turning point in the early detection, categorization, and classification of diseases. The authors aim to find a solution to the problem of detection of early WPW using ECG images with the help of deep learning models. Dataset extraction of positive and negative WPW ECG images, deep learning models implementation, and comparison of the achieved results has been done in the current paper. Models implemented for the study are CNN, which specializes in image processing, and LeNet, and MobileNet, which are different variants of CNN itself. After iterative hyperparameter tuning, CNN was found to achieve the best accuracy among all the three tested models. The authors, other than combining two great fields to solve a serious problem, also aim to provide a helpful study for medical researchers and practitioners and motivate individuals to take good care of themselves and indulge in regular check-ups for early detection of such diseases.